18 results about "Posterior probability density" patented technology

The invention relates to a seismic liquefaction assessment method based on conditional random field simulation, which comprises the following steps: firstly, obtaining a logarithmic normal distribution random field of a target area under a corresponding SPT-N value, then resampling through a Bootstrap method, constructing a weighted prior probability density function of the target area in combination with a likelihood function, and finally calculating the seismic liquefaction of the target area according to a Bayesian theory. A Markov chain Monte Carlo sampling method is combined, through posterior probability density distribution, an optimal horizontal direction correlation distance is determined, a covariance matrix is constructed to generate a conditional random field, and then through multiple times of simulation, the conditional random field is converged; and finally, aiming at the target area, through calculation of a cyclic stress ratio and a cyclic resistance ratio, constructing a liquefaction probability distribution diagram corresponding to the target area. According to the method, a conditional random field simulation method is inferred and improved by combining Bootstrap and Bayesian theories, the precision and reliability of geological parameter simulation are remarkably improved, and reliable data support is provided for seismic liquefaction assessment of deep and uneven site engineering.

The invention relates to a method and apparatus for single epoch position bounding. A method for determining a protection level for a position estimate using a single epoch of GNSS measurements, the method comprising: specifying a prior probability density of states x P(x); specifying a system model h(x) relating states x to measured observations z; quantifying a quality metric q associated with the measurements; specifying a non-Gaussian residual error probability density model f(r|θ,q) and fitting the model parameters θ using a set of experimental data; defining a posterior probability density P(x|z,q,θ); estimating states x; and calculating the protection level by integrating the posterior probability density P(x|z,q,θ) over states x.

This invention relates to the field of electromagnetic exploration technology, and discloses an efficient probabilistic inversion method, medium, and electronic equipment for electromagnetic resource exploration. The method employs a Fourier neural network from machine learning, training it with a large number of conductivity models to obtain the electromagnetic forward response. This network, through a cascaded structure of multi-layer neurons, automatically extracts features from a large amount of magnetotelluric response data, constructing a mapping from conductivity models to magnetotelluric responses. This avoids traditional forward calculations for conductivity models obtained from each sampling, accelerating the forward calculation time during sampling. Prior constraints related to the structure are constructed in advance and added to the inversion prior distribution, reconstructing the Metropolis-Hasting acceptance criterion. The prior information is used to quickly search for the posterior probability density distribution of the model, reducing the search time in the inversion space during sampling, and quantitatively evaluating the obtained inversion results.

The invention discloses a structure load and parameter joint identification method based on Gibbs sampling. The method comprises the following steps: S1, establishing a discrete state space model of a linear time-invariant dynamic system; s2, fitting an unknown external load by adopting an orthogonal polynomial, and representing the external load as a linear combination of an orthogonal polynomial basis function and a fitting coefficient; s3, defining an uncertainty parameter set, wherein the set comprises the fitting coefficient, the measurement noise, the unknown structure parameter and the fitting error; s4, constructing a multilayer Bayesian model based on the Bayesian theory, wherein the model comprises a likelihood function layer, a load prior model layer and hyper-parameters; s5, solving the multilayer Bayesian model by adopting a Gibbs sampling method, and obtaining posterior probability density distribution of each uncertainty parameter through iterative sampling; and S6, based on the posterior probability density distribution, obtaining an identification result of the external load and the unknown structure parameters. According to the method, the parameter uncertainty is considered, and the recognition precision and robustness of the dynamic load and parameters of the complex structure are improved.

The present application relates to radar target tracking technical field, specifically to a kind of for radar weak target's detection front tracking method.For simultaneously estimating target state and measurement noise covariance, first need to calculate the joint probability density of target state and measurement noise covariance.Then introduce Gaussian inverse gamma mixture distribution to model joint probability density, since target state and measurement noise covariance are coupled in joint likelihood function, this will lead to joint posterior probability density difficult to solve analytically, therefore, approximate solution of separable approximation of joint posterior probability density is solved using variational bayesian method.Finally, in filtering update phase, based on the separable approximation solution, information exchange is performed, i.e., each Bernoulli component uses the predicted state information shared by other Bernoulli components to perform update.Simulation verification shows that, in low signal-to-noise ratio scenario, the present application can adaptively estimate measurement noise covariance, and tracking accuracy is improved.

The present invention proposes an independent fusion modeling and response prediction method for low- and high-precision missile performance data. The method utilizes collected missile reliability test data of different accuracies to respectively construct a low-precision response model and a high-precision response model. Based on Bayesian theory, the posterior probability density distribution of each model parameter, and the high-precision response model, the posterior probability density function of the high-precision missile performance response data at the influencing factor data of any application scenario is obtained, and then a high-precision missile performance response data prediction model at the influencing factor data of any application scenario is obtained. The method can predict the high-precision missile performance response data at the influencing factor data of any application scenario, and ensure the accuracy of the response prediction result and the efficiency of the solution operation.

The invention discloses a method for performing extended target tracking by using a random matrix and partial normal distribution, which comprises the following steps of: establishing an evolution model and a partial normal distribution measurement model, and performing prior prediction on a prior target motion state, an extended form and a measurement deflection variable; carrying out posteriori estimation on a target motion state, an expansion form and a measurement skew variable, and obtaining a skew constraint vector estimation value; re-modeling based on an IT-IMM framework: obtaining an optimal solution of a prior probability density function of a target motion state, an extension form and a measurement skew variable through a weighted KLA algorithm; the posterior probability density of the corresponding mode is obtained through a variational Bayesian algorithm, and mode probability updating is carried out; and through weighted KLA approximation, final estimation of the motion state and the expansion form of the target is obtained. According to the method, accurate tracking of the target motion state and accurate estimation of the expansion form can be realized.

This invention discloses a real-time probabilistic inversion method for TBM (Tunnel Boring Machine) extrusion load based on a surrogate model. First, the extrusion load vector and its prior value range are defined for the TBM shield tunneling environment. A surrogate model is constructed and iteratively updated based on a small number of high-fidelity numerical simulation samples to establish a rapid mapping relationship between the extrusion load and the shield structure response. Then, on-site measured shield monitoring data are acquired, and a likelihood function is constructed by combining sensor observation errors and surrogate model prediction errors. Finally, within a Bayesian probabilistic inversion framework, a Markov chain-Monte Carlo algorithm is used to call the surrogate model for sampling, obtaining the posterior probability density distribution, optimal estimate, and confidence interval of the extrusion load, and calculating the shield structure failure probability accordingly. This invention can meet the real-time requirements of inversion calculations and provide a quantitative assessment of the uncertainty of the inversion results, offering a scientific basis for safety decisions when TBMs traverse strata with large extrusion deformation.

A method for determining a protection level of a position estimate using a single epoch of GNSS measurements, the method includes: specifying a prior probability density P(x) of a state x; specifying a system model h(x) that relates the state x to observables z of the measurements; quantifying quality metrics q associated with the measurements; specifying a non-Gaussian residual error probability density model ƒ(r|θ, q) and fitting model parameters θ using a set of experimental data; and defining a posterior probability density P(x|z, q, θ); estimating the state x; and computing the protection level by integrating the posterior probability density P(x|z, q, θ) over the state x.

The invention discloses a current transformer error state identification method and system based on a dual-target coevolution multi-chain MCMC, and the method comprises the steps: constructing a Bayesian multiple linear regression model of a transformer area load current and a user side load current according to the secondary side current data of a transformer area CT and the secondary side current data of a user side CT which are in online operation; obtaining a corresponding complex regression coefficient under the maximum posterior probability; based on a dream optimization mutation strategy, according to the real part posterior probability density and the imaginary part posterior probability density of the complex regression coefficient, obtaining a modulus and an argument of the optimal complex regression coefficient through a dual-objective co-evolution multi-chain MCMC algorithm; and comparing the modulus statistic and the argument statistic with corresponding control thresholds respectively, and performing error state identification on the current transformer according to a comparison result. According to the method, accurate identification of the abnormal state of the current transformer is realized through a Bayesian multiple linear regression model, a dual-target coevolution multi-chain MCMC algorithm and a dream optimized variation strategy.

The invention provides a distributed multi-target fusion tracking method and device based on time calibration, and relates to the technical field of sensors. The distributed multi-target fusion tracking method based on time calibration comprises the following steps: on the basis of a generalized covariance cross fusion criterion, constructing a posterior probability density expression of a time calibration parameter by using a Gaussian mixture approximation technology and a probability hypothesis density filter; selecting a mean value of the Gaussian component with the maximum weight from a plurality of Gaussian components of the posterior probability density expression as an estimated value of the time calibration parameter; and based on a generalized covariance cross fusion criterion, fusing the first multi-target probability density function and the calibrated second multi-target probability density function to obtain a multi-target state estimation value of the tracking target. According to the invention, data from a plurality of sensor nodes can be efficiently and accurately fused in a distributed multi-target tracking scene so as to realize accurate positioning of a tracking target.

The invention belongs to the technical field of power system fault detection, and particularly relates to an equipment fault detection method and system based on a multi-state Markov chain, and the method comprises the steps: employing a directed graph model to represent a power distribution network structure, building a transition probability matrix based on the defined directed graph model, and calculating the prior probability density; and calculating posterior probability density based on the prior probability density by adopting a Bayesian algorithm in combination with real-time observation data, sampling from the posterior probability density by utilizing an MCMC algorithm to obtain posterior probability density estimation, and correcting the posterior probability density estimation by utilizing Gaussian process regression to be used for state estimation so as to realize fault detection of the power distribution equipment. According to the method, on one hand, observation data obtained by a sensor in real time are dynamically fused when the prior probability density is calculated by adopting the Bayesian algorithm, so that the model adaptability is improved, and on the other hand, the posterior probability density estimation is corrected by utilizing the residual error of Gaussian process regression prediction, so that the online fault detection precision is improved.

The application relates to a method and device for establishing an initial model of seismic inversion based on inverse mapping of lithofacies probability in the field of oil and gas exploration. The method for establishing an initial model of seismic inversion based on inverse mapping of lithofacies probability comprises the following steps: collecting and processing data to obtain post-stack seismic data and a lithofacies curve; obtaining a plurality of seismic attribute bodies from the post-stack seismic data, and combining the lithofacies curve to obtain a pseudo well curve; training a neural network model by using the pseudo well curve, and outputting a lithofacies classification result and a lithofacies probability body; analyzing the lithofacies classification result and the lithofacies probability body to establish a posterior probability density space of physical property parameters; and based on the posterior probability density space and a probability density distribution, using a Monte Carlo algorithm and a rejection-acceptance sampling to generate a physical property parameter combination meeting a condition, and obtaining the initial model of seismic inversion. The application can more accurately perform reservoir characterization, and the consistency between the initial model of seismic inversion and a well point is good.

The application discloses a kind of original light element abundance prediction and parameter estimation method, including setting cosmological parameter and the sampling range of rigid phase correction factor, batch generation structured data set containing multiple light element theoretical abundance using numerical calculation code;Deep network model integrating multi-head attention mechanism and deep residual block is constructed, and the nonlinear mapping from parameter space to abundance space is established by supervised training;The model is encapsulated as a likelihood function interface, integrated into the Cobaya Bayesian analysis framework, and the forward inference is performed using the computing kernel;Combined with the actual observation data constraint, the parameter space is automatically sampled using the Markov Chain Monte Carlo algorithm, and the posterior probability density distribution of each physical parameter is output.The application effectively solves the problem of low efficiency in solving high-dimensional rigid equation set in original nucleosynthesis, and can quantitatively analyze the tension between lithium abundance observation data, and reveal the potential physical path of rigid phase correction parameter to alleviate the lithium problem.

The invention relates to the technical field of electromagnetic prospecting, and discloses a resource electromagnetic prospecting efficient probability inversion method, a medium and electronic equipment, the method adopts a Fourier neural network in machine learning, an electromagnetic forward modeling response is obtained through training by inputting a large number of conductivity models, and the network is connected with the electromagnetic forward modeling response through a cascade structure of multiple layers of neurons. Features are automatically extracted from a large amount of magnetotelluric response data, mapping from a conductivity model to magnetotelluric response is constructed, traditional forward modeling calculation of the conductivity model obtained through sampling each time is avoided, and the forward modeling calculation time in the sampling process is shortened; according to the method, the prior constraint of a related structure is constructed in advance and is added into inversion prior distribution, a Metropolis-Hasting acceptance criterion is reconstructed, posterior probability density distribution of a model is quickly searched by using prior information, the search time in an inversion space during sampling is shortened, and quantitative evaluation is performed on an obtained inversion result.

The invention provides a multi-source cross-scale satellite data adaptive greenhouse gas product fusion method, which belongs to the technical field of atmospheric environment remote sensing application, and comprises the following steps: carrying out space-time unified matching, resampling and reprojection on observation data of different satellite platforms; performing error correction by using high-resolution OCO-2 data as hard data, performing system deviation correction on low-resolution GOSAT data based on an XGBoost algorithm, and constructing soft data; the hard data and the soft data are fused through a Bayesian maximum entropy framework, the fusion weight is optimized by utilizing a space-time covariance model and soft data constraints, and a posterior probability density mean value is obtained to serve as a final fusion result. And a high-precision CO2 concentration field in a global range is obtained by combining a space-time large-scale trend. According to the method, the contribution of low-uncertainty observation can be enhanced, high noise and geometric mismatch interference are effectively suppressed, a global greenhouse gas concentration fusion product with multi-source information and an uncertainty field is generated, and the precision and space continuity of greenhouse gas monitoring are remarkably improved.

The invention provides a distributed target fusion tracking method and device based on space-time registration, and relates to the technical field of sensors. The distributed target fusion tracking method based on space-time registration comprises the following steps: selecting a mean value of a Gaussian component with the maximum weight from a plurality of Gaussian components of a posterior probability density expression of a space-time registration parameter as an estimated value of the space-time registration parameter; aligning the second posterior probability density with the first posterior probability density in time and space according to an estimated value of a space-time registration parameter by using a space-time registration formula of a uniform-speed motion model to obtain a calibrated second posterior probability density; and based on a generalized covariance cross fusion criterion, fusing the first posterior probability density and the calibrated second posterior probability density to obtain a target existence probability and a target space density of the tracking target. According to the invention, registration of multi-node sensor data can be realized with high precision and high reliability.

The invention particularly relates to a Gaussian mixture distribution homomorphic encryption fusion method based on key information packaging. The method comprises the following steps: establishing a state equation of target motion and a sensor measurement model; predicting and updating a target state by using a Gaussian mixture probability hypothesis density filter to obtain posterior probability density distribution of each sensor node, and performing pruning and merging processing to obtain a target posterior estimation Gaussian set; converting a mean value and a covariance in the posteriori estimation Gaussian set of each sensor node into an information vector and a precision matrix; uniformly quantizing the information vector and the precision matrix, and encrypting quantized data through a Paillier homomorphic encryption method; and transmitting the encrypted data to a fusion center for geometric mean fusion, and carrying out decryption and inverse quantization on the encrypted data by the fusion center to obtain a state estimation result of the target. According to the method, under the condition that the original fusion precision is not lost, the leakage risk of key information in the fusion process is further reduced.